As kalniel and Tarinder already said: what scene is being benchmarked?
Skyrim (or any Elder Scrolls Game) does not have a built-in benchmark. Neither does Fallout 4. So what people are generally doing is using a save game and repeating a certain look around (for example Skyrim Dragonreach view-around) or moves. Totally dependent on where this save game is though, so not ideal for comparing sites but should be roughly repeatable for benchmark runs.
Actually for Skyrim, something like the intro carriage ride or a similar 'cut scene' would probably be better except that the intro is not taxing.
While I appreciate what you're saying and I do realise they have advantages in VR and power draw. In terms of raw performance / pound they're significantly worse than my GTX970 purchase. All that matters is performance today and price today and in that regard they're way off the mark. What difference does it make to my experience when a card was released if they perform roughly on par?
I get what you're saying, but I think it's just a bit disingenuous to compare the just launched price to the established price 12 months after release. When you bought your card the 480 wasn't even available so it couldn't be compared. Now, for a very limited time, they overlap. In the not too distant future you won't be able to buy old 970 stock, and the 480 will drop in price when stock arrives in volume. But yes - if you have to buy a card today, then the Strix 480 doesn't make a good value argument (unless you are considering freesync or dx12 of course). Of course, a second hand card might given even better still value.
I think the major disadvantage of getting a 970/980 right now is: if you keep things a long time, then you might get to the end of mainline driver support as you are already 2 years into the roughly 5 year support cycle.
Indeed, the absolute best air-cooled OC results I saw, with a HUGE aftermarket cooler strapped on, only reached 1450MHz. 1500MHz was only acheivable with a custom water loop. 1500Mhz is probably achievable, but it's not going to be easy or straightforward, and you certainly won't get any board partner guaranteeing it out of the box...
Hexus is paid by Nvidia, that is why they use the old driver for Tomb Raider that runs ALL AMD cards like trash, when the new patch ups their performance by over 20% percent across the board, so AMD cards become as fast as the competition.
They also run several of the games in DX11 when DX12 is available and still run the terrible and old openGL driver, when Vulkan is available.
They also don't test games like Dota 2 in Vulkan and Assasins Creed games who use newer technology for rendering of their games.
Want to provide any evidence to back up that statement, or do you just enjoy randomly slandering companies on forums they own?
Tarinder has already stated in previous reviews that hexus are in the process of updating their benchmark selection. They only do that once every couple of years so that you can compare reviews between generations in a useful way (if every review used a different set of benchmarks you couldn't do any comparisons without rebenchmarking ALL cards, and that is prohibitively time consuming). As it is, I hardly think Doom in Vulkan or Hitman are benchmarks that are designed to make the RX 480 look bad, given it trounces the GTX 1060 in both those benchmarks; neither are The Division or Total War: Warhammer, in which the performance is even overall. Given that the RX 480 is generally accepted to be slower than the GTX 1060 across a wide range of games (TPU run around 15 games and found the 1060 to be overall ~ 6% faster), Hexus using 3 games where the GTX 1060 is faster, 2 where the RX 480 is faster, and 2 where they're about equal, is probably fair. They also show the OpenGL figures for Doom, sure, and that loks bad for AMD because their OpenGL drivers are just that awful. There's still a few OGL-only games out there, and if you're playing those games an RX 480 is going to suck because AMD's drivers are bad. I think that's also a valid point to make in a review - particularly when it's balanced by a demonstration of how good the RX 480 is in Vulkan for that title.
Horses for courses. Hexus generally have a fairly well balanced review. The GTX 1060 is almost certainly a technically better card than the RX 480. I think you'll have a hard time proving that Hexus' review is particularly biased towards nvidia. At which point, it's probably a bad idea to accuse them of deliberately favouring nvidia for financial reasons. That's a pretty serious accusation...
I've been waiting for this review - thanks!
Were the temperatures achievd with the same fan speeds or were the RX480s fans spinning faster?
Did the cards produce any coil whine?
There are currently 1 users browsing this thread. (0 members and 1 guests)